272 
hm? 
if cvis here = - 
—. We can, ina way similar to that used by vAN DER 
x 
Waars Jr. in solving the integral developed in his paper, evaluate G 
in terms of a series by expanding e°®% —e °/ into a series of ascending 
powers of g. The proof of convergence can be given in the same 
way as by vAN DER WaAALs Jr. 
If we write 
Me” 
OS gait ia ne 
for the potential energy of two molecules in contact with the axes 
of their doublets parallel and at the same time perpendicular to the 
line joining their centres and if we take as our upper limit r= 
we obtain 
| Ne: See 1 Ber. ; 
iB = Aa 0 EN q, (Avy? — 55) gq, (hv)* — | 
ER Se \ 
ERN (ey — NON fy 
in which 
Mes 
bm te ee —— 
Se ey wel are aa = 
I, = Ata + Tik Es 
Ee 
WT al Me gat AE enn 
1 EN 1 
Pay etre et Pia mee MN soe 
3 5 7 
or 
PR =. 
: Piet Heat oe ie a = Fe) = zes (*) we (59) 
1 
in which h should be replaced by a if B is to be obtained as a 
= 
function of 7. We now obtain a series containing only the even 
powers of T—! (ef. § 5). 
Just as in § 5 we can now separate the terms which represent 
the collision virial and the attraction virial in 5. 
Should the law of dependence of B upon temperature for a dia- 
tomie substance in the region for which y,,4=°/, £ *) be found expe- 
1) In Communication N°. 1094 § 7 (March ’09) the dependence of B upon tem- 
perature is given for hydrogen as deduced from the isotherms of KAMERLINGH 
Onnes and BRAAK. With regard to the specific heat, however, we must remark 
that measurements made by Eucken, Berlin Sitz-Ber., Febr. 1912, p. 141, of the 
